INDEX

Explanations

references to governmental and political entities associated with influence and funding

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

reso

-0.17

apel

-0.15

ault

-0.15

 lives

-0.14

Tests

-0.14

rey

-0.14

imp

-0.14

 Lives

-0.13

vale

-0.13

 conserve

-0.13

POSITIVE LOGITS

 paid

0.19

 influence

0.18

 Influence

0.18

 launder

0.18

 smear

0.17

sock

0.17

 propag

0.17

 malign

0.17

 PROP

0.16

 troll

0.16

Activations Density 0.038%

references to governmental and political entities associated with influence and funding

No Comments

No Known Activations